session/statistics: discard feedbacks from delete / update (#17452)…

… (#17840)

* cherry pick #17452 to release-2.1

Signed-off-by: sre-bot <sre-bot@pingcap.com>

* resolve conflicts

Co-authored-by: Kenan Yao <cauchy1992@gmail.com>
Co-authored-by: ti-srebot <66930949+ti-srebot@users.noreply.github.com>

executor/builder.go

@@ -1797,7 +1797,10 @@ func buildNoRangeTableReader(b *executorBuilder, v *plannercore.PhysicalTableRea
 	} else {
 		e.feedback = statistics.NewQueryFeedback(e.physicalTableID, ts.Hist, int64(ts.StatsCount()), ts.Desc)
 	}
-	collect := e.feedback.CollectFeedback(len(ts.Ranges))
+	collect := statistics.CollectFeedback(b.ctx.GetSessionVars().StmtCtx, e.feedback, len(ts.Ranges))
+	if !collect {
+		e.feedback.Invalidate()
+	}
 	e.dagPB.CollectRangeCounts = &collect
 
 	for i := range v.Schema().Columns {
@@ -1854,7 +1857,10 @@ func buildNoRangeIndexReader(b *executorBuilder, v *plannercore.PhysicalIndexRea
 	} else {
 		e.feedback = statistics.NewQueryFeedback(e.physicalTableID, is.Hist, int64(is.StatsCount()), is.Desc)
 	}
-	collect := e.feedback.CollectFeedback(len(is.Ranges))
+	collect := statistics.CollectFeedback(b.ctx.GetSessionVars().StmtCtx, e.feedback, len(is.Ranges))
+	if !collect {
+		e.feedback.Invalidate()
+	}
 	e.dagPB.CollectRangeCounts = &collect
 
 	for _, col := range v.OutputColumns {
@@ -1928,10 +1934,13 @@ func buildNoRangeIndexLookUpReader(b *executorBuilder, v *plannercore.PhysicalIn
 	} else {
 		e.feedback = statistics.NewQueryFeedback(e.physicalTableID, is.Hist, int64(is.StatsCount()), is.Desc)
 	}
-	// do not collect the feedback for table request.
+	// Do not collect the feedback for table request.
 	collectTable := false
 	e.tableRequest.CollectRangeCounts = &collectTable
-	collectIndex := e.feedback.CollectFeedback(len(is.Ranges))
+	collectIndex := statistics.CollectFeedback(b.ctx.GetSessionVars().StmtCtx, e.feedback, len(is.Ranges))
+	if !collectIndex {
+		e.feedback.Invalidate()
+	}
 	e.dagPB.CollectRangeCounts = &collectIndex
 	if cols, ok := v.Schema().TblID2Handle[is.Table.ID]; ok {
 		e.handleIdx = cols[0].Index

statistics/feedback.go

@@ -88,16 +88,18 @@ var (
 )
 
 // CollectFeedback decides whether to collect the feedback. It returns false when:
-// 1: the histogram is nil or has no buckets;
-// 2: the number of scan ranges exceeds the limit because it may affect the performance;
-// 3: it does not pass the probabilistic sampler.
-func (q *QueryFeedback) CollectFeedback(numOfRanges int) bool {
+// 1: the feedback is not generated by select query;
+// 2: the histogram is nil or has no buckets;
+// 3: the number of scan ranges exceeds the limit because it may affect the performance;
+// 4: it does not pass the probabilistic sampler.
+func CollectFeedback(sc *stmtctx.StatementContext, q *QueryFeedback, numOfRanges int) bool {
+	if !sc.InSelectStmt {
+		return false
+	}
 	if q.hist == nil || q.hist.Len() == 0 {
-		q.Invalidate()
 		return false
 	}
 	if numOfRanges > MaxNumberOfRanges || rand.Float64() > FeedbackProbability {
-		q.Invalidate()
 		return false
 	}
 	return true

statistics/update_test.go

@@ -1445,3 +1445,33 @@ func (s *testStatsUpdateSuite) TestUnsignedFeedbackRanges(c *C) {
 		c.Assert(tbl.Columns[1].ToString(0), Equals, tests[i].hist)
 	}
 }
+
+func (s *testStatsUpdateSuite) TestDeleteUpdateFeedback(c *C) {
+	defer cleanEnv(c, s.store, s.do)
+	testKit := testkit.NewTestKit(c, s.store)
+
+	oriProbability := statistics.FeedbackProbability
+	defer func() {
+		statistics.FeedbackProbability = oriProbability
+	}()
+	statistics.FeedbackProbability = 1
+
+	h := s.do.StatsHandle()
+	testKit.MustExec("use test")
+	testKit.MustExec("create table t (a bigint(64), b bigint(64), index idx_ab(a,b))")
+	for i := 0; i < 20; i++ {
+		testKit.MustExec(fmt.Sprintf("insert into t values (%d, %d)", i/5, i))
+	}
+	c.Assert(h.DumpStatsDeltaToKV(statistics.DumpAll), IsNil)
+	testKit.MustExec("analyze table t with 3 buckets")
+
+	testKit.MustExec("delete from t where a = 1")
+	c.Assert(h.DumpStatsDeltaToKV(statistics.DumpAll), IsNil)
+	c.Assert(len(h.GetQueryFeedback()), Equals, 0)
+	testKit.MustExec("update t set a = 6 where a = 2")
+	c.Assert(h.DumpStatsDeltaToKV(statistics.DumpAll), IsNil)
+	c.Assert(len(h.GetQueryFeedback()), Equals, 0)
+	testKit.MustExec("explain analyze delete from t where a = 3")
+	c.Assert(h.DumpStatsDeltaToKV(statistics.DumpAll), IsNil)
+	c.Assert(len(h.GetQueryFeedback()), Equals, 0)
+}